R 语言 copula - CoVaR 程序分享

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

这是最近整理的关于 copula - CoVaR 的程序， 有兴趣在这个领域交流的，可以加好友。

参考论文：
Federal Reserve Bank of New YorkStaff Reports
CoVaR
Tobias AdrianMarkus K. Brunnermeier
Staff Report No. 348September 2008Revised September 2014

程序代码：

1. # CoVaR value evaluting at quantile CoVaR and quantile VaR.
   
2. 
   
3. # Description
   
4. #Calculate the conditional quantile or CoVaR with different type of Copula and marginal distribution. In this package several bivariate copula families are included for bivariate analysis.
   
5. #It provides functionality of elliptical (Gaussian and Student t) as well as Archimedean (Clayton, Gumbel, Frank, Plackett, BB1, SCJ, rotated clayton and rotated Gumbel) copulas to cover a large bandwidth of possible dependence structures.
   
6. 
   
7. 
   
8. 
   
9. # Author
   
10. #Andrea Ugolini <andreaugolini@me.com> \\
    
11. #Juan Carlos Reboredo Noguiera <juancarlos.reboredo@usc.es>
    
12.   
    
13.   # References
    
14. #  Reboredo, J. C., & Ugolini, A. (2016).
    
15. #Quantile dependence of oil price movements and stock returns. Energy Economics, 54, 33-49.
    
16. 
    
17. 
    
18. # Section: Example --------------------------------------------------------
    
19. rm(list=ls())
    
20. graphics.off()
    
21. setwd(dirname(rstudioapi::getActiveDocumentContext()$path))
    
22. 
    
23. # Example
    
24. #### RCoVaRCopula
    
25. load("Data_demo.Rdata")
    
26. source("CoVaR.R")
    
27. source("DynCopulaCoVaR.R")
    
28. source("DynCopulaCoVaRUpper.R")
    
29. source("skewtdis_inv.R")
    
30. require("pracma")
    
31. require("copula")
    
32. 
    
33. #### CoVaR Downside
    
34. 
    
35. CoVaR1part = CoVaR(0.05,0.05,par=par1_1,par2=par2_1,dof=tailBrazil,gamma=asyBrazil,cond.mean=meanBrasil1,
    
36.                    cond.sigma=sigmaBrasil1,dist="tskew",type="Student")
    
37. 
    
38. CoVaR2part = CoVaR(0.05,0.05,par=par1_2,par2=par2_2,dof=tailBrazil,gamma=asyBrazil,cond.mean=meanBrasil2,
    
39.                    cond.sigma=sigmaBrasil2,dist="tskew",type="Student")
    
40. #### CoVaR Upside
    
41. 
    
42. CoVaR1partUp = CoVaR(0.95,0.95,par=par1_1,par2=par2_1,dof=tailBrazil,gamma=asyBrazil,cond.mean=meanBrasil1,
    
43.                      cond.sigma=sigmaBrasil1,dist="tskew",type="StudentUp")
    
44. 
    
45. CoVaR2partUp = CoVaR(0.95,0.95,par=par1_2,par2=par2_2,dof=tailBrazil,gamma=asyBrazil,cond.mean=meanBrasil2,
    
46.                      cond.sigma=sigmaBrasil2,dist="tskew",type="StudentUp")
    
47. 
    
48. CoVaR1D = CoVaR1part$CoVaR
    
49. CoVaR2D = CoVaR2part$CoVaR
    
50. CoVaR1U = CoVaR1partUp$CoVaR
    
51. CoVaR2U = CoVaR2partUp$CoVaR
    
52. 
    
53. TimeCoVaRD = rbind(CoVaR1D,CoVaR2D)
    
54. TimeCoVaRU = rbind(CoVaR1U,CoVaR2U)
    
55. 
    
56. #### Plot
    
57. plot(as.matrix(TimeCoVaRD),type="l",col="blue",
    
58.      ylim=c(-0.5,0.5),xlab="Time",ylab="")
    
59. lines(VaR,col="black",lty=2)
    
60. lines(TimeCoVaRU,col="red",lty=4)
    
61. lines(VaRup,col="green",lty=3)
    
62. abline(h=0,col="gray33")

复制代码

绘图：



联系方式： 2974861304

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏1 回帖

关键词：Copula opula CoVar VaR Brunnermeier

AndySims 是我去年申请的辅助号；
这个号今天才能登录，  copula - TGARCH - CoVaR, 可以参考 我2019年分享论坛的这个帖子；不过后来修改过很多地方，论坛的代码与最新的不一样，这个帖子里的主程序，就使用了一些修改的地方；

https://bbs.pinggu.org/thread-7523839-1-1.html

以后可能，Andy 这个号会较少登录；

手机掉了， 大号QQ无法登录 （280201722） ，
临时 QQ ： 2974861304